CN111898096B

CN111898096B - System and method for analyzing and visualizing national transition of football player

Info

Publication number: CN111898096B
Application number: CN202010717856.6A
Authority: CN
Inventors: 巫英才; 曹安琪
Original assignee: Zhejiang University ZJU
Current assignee: Zhejiang University ZJU
Priority date: 2020-07-23
Filing date: 2020-07-23
Publication date: 2024-02-09
Anticipated expiration: 2040-07-23
Also published as: CN111898096A

Abstract

The invention relates to a system and a method for analyzing and visualizing a cross-country transfer of football players, belonging to the technical field of data analysis and visual analysis. Comprising the following steps: a national view for displaying transnational transfer strategies for each tournament in the intercontinental football league and the country/region for the expert to browse, select and further analyze; an influence view, providing a customized visual component for an expert to observe the change of the bidirectional influence intensity and further analyze the causal relationship; and adjusting the view, providing an interaction method for an expert to adjust the quantity and quality of the nationwide meeting players and obtaining a simulation result. The method can analyze the influence intensity between the national transfer of football players and the national team performance in different countries, helps the expert to find the effect and influence of the football player transfer policy, and is convenient for the subsequent establishment of the appropriate football player national transfer policy.

Description

System and method for analyzing and visualizing national transition of football player

Technical Field

The invention relates to the technical field of data analysis and visual analysis, in particular to a system and a method for analyzing and visualizing a cross-country transfer of football players.

Background

Football is deeply favored by wide sports lovers, and is one of the most popular sports in the world. The national transition of professional football players has also become a non-negligible phenomenon in the football field. In recent years, the problem of how a player's transnational meeting affects national team performance, and how national team performance affects a player's transnational meeting has attracted considerable interest to researchers. A national transfer of players represents a phenomenon in which players play in a tournament outside the country or region in which they are located, including the direction in which two players flow, namely a player import and a player export. Specifically, most countries and regions of the world have organized domestic tournaments, each of which contains many clubs. The club is allowed to trade players from different countries and in the process promotes bi-directional transnational transfer of players.

In order to be able to make appropriate decisions about the player's transnational meeting policy, researchers have proposed many data-driven approaches. Among them, statistical analysis, particularly regression-based analysis, has been widely used to evaluate the impact between a player's transnational meeting and national team performance. However, the impact between the team's transnational transition and the national team performance is bi-directional and complex to analyze. The domain expert needs causal relationships between the two-way effects to fully analyze the impact process. However, existing regression-based solutions are limited to only one-way impact of players at transnational gatherings on national team performance, ignoring causal relationships. Such a traditional regression model would prevent a full understanding of the impact of a player's transnational transition. Meanwhile, domain experts face challenges in deeply analyzing and interpreting the results of the above analysis model. In the case of developing a transnational transfer policy, domain experts need to quickly browse raw data to select model inputs and interact with the model based on domain knowledge. Thus, domain experts find it difficult to utilize statistical analysis methods in the case of practical applications to obtain valuable insights that can be applied in decision-making scenarios.

Visual analysis technology support experts interact and interpret to integrate domain knowledge into the analysis process, thereby adjusting the model and exploring detailed player information to improve understanding of the results. However, in the field of sports visual analysis, most of the work mainly focuses on detailed score table data and trajectory data for game result analysis and tactical analysis. There is still a lack of visual analysis methods tailored to the player's transnational transition. At the same time, the visualization of regression models also has little attention to the relationship between nationwide transitions and national team performance. In view of the unique data structure and field tasks, it is difficult to directly apply current research to football turn analysis.

Therefore, there is a need for a visual method of soccer player transnational transition analysis that helps field experts explore the bi-directional impact between player transnational transition and national team performance and draw conclusions from a large number of complex soccer player transition data.

Disclosure of Invention

The invention aims to provide a system and a method for analyzing and visualizing cross-country transfer of football players, which can analyze the influence intensity between cross-country transfer of football players in different countries and national team performances, help experts find the effect and influence of the cross-country transfer policy of football players, and facilitate the subsequent establishment of a proper cross-country transfer policy of football players.

To achieve the above object, in a first aspect, the present invention provides a system for cross country transfer analysis visualization of football players, comprising:

a national view for displaying transnational transfer strategies for each tournament in the intercontinental football league and the country/region for the expert to browse, select and further analyze;

an influence view, providing a customized visual component for an expert to observe the change of the bidirectional influence intensity and further analyze the causal relationship;

and adjusting the view, providing an interaction method for an expert to adjust the quantity and quality of the nationwide meeting players and obtaining a simulation result.

Preferably, the country view includes a map component for displaying a global player transnational transfer overview and a statistical world map and a projected similarity component for providing a global transnational transfer strategy in terms of similarity.

The map component displays global player transnational meeting summaries as well as statistical world maps for users to browse intercontinental football leagues of interest. Because the same continental transnational meeting policies are generally similar, users typically choose a country from a single intercontinental football league (e.g., the soccer Union), analyze it, and choose the intercontinental football league based on detailed transnational meeting information. A statistical world map is provided in the map component to show the values of the transnational transfer variables. The user can switch the cross country transfer variables using a button above the statistical world map. Opacity is used in the map component to encode particular values of variables of the import or export players for the selected country.

The similarity component provides a projection of global transnational transfer policies in terms of similarity. The user needs to analyze countries with policies that are similar to transnational transitions. Therefore, a scatter plot is employed in the similarity component to show how similar all nationwide transfer policies are. The location of the points in the scatter plot is determined by the t-SNE projection algorithm. In the t-SNE projection, the original vector of a country consists of the total number of players in the transnational transition, the average market value, the average number of goals and the average goals, these averages being variables often used in the transnational transition analysis of football players and typically used to describe the transnational transition strategy. By projecting the result, the user can distinguish countries that are strategically similar in the transnational transfer. The opacity is used in the scatter plot to encode the transnational transition variables for each country.

The transnational transfer variables include: the number of players, the price of the players, the average number of times of the players in single season and the average number of goals in the single season.

Preferably, the country view includes the following interactions:

time screening, wherein a user filters an interested time interval by dragging a sliding block;

a direction switch, wherein the user displays an import player or an export player in the national view through button switch;

country searching, wherein a user searches for a country of interest through a search bar;

the user selects an intercontinental football league in the map component or a country in the similarity component by clicking or framing the selected intercontinental football league or country to be highlighted in color.

Preferably, the influence view includes an influence graph that displays influence intensities over years by placing a bar graph with a timeline; the height of the bar graph represents the impact strength calculated by the cross-hysteresis path analysis model.

Preferably, in the detailed impact mode, circles are used to encode the primary impact direction between the player variable of the transnational meeting and the national team performance, the colors of the circles encode the primary impact direction, and the areas encode the difference in impact intensity of the two directions.

The effect map may be visualized by crossing the results of the lagged path analysis model. Comparing impact strengths by year is the primary task in the impact view. Thus, for direct comparison, the influence graph places the bar graph directly with the timeline to show the influence intensity over the years. The height of the bar graph represents the impact strength calculated by the cross-hysteresis path analysis model (the higher the height, the greater the impact strength). In order to clearly show the trend of variation affecting intensity, an exponential scaling option of bar chart height is provided in addition to the linear scaling. The influence map can be switched between a total influence and a detailed influence. Under the total influence mode, the user can compare the total influence intensity of the imported player and the exported player and simultaneously find the change trend. Circles are used in the detailed impact mode to encode the primary impact direction between the player variable and the national team performance of the transnational meeting. The color of the circle codes the main influencing direction and the area codes the difference of the influencing intensities of the two directions (the larger the area, the larger the influencing intensity difference). The circles in the effect map are aligned with the corresponding bar graphs to facilitate finding causal relationships of the two-way effect over time.

Preferably, the influence view includes the following interactions:

merging years, brushing the years in an influence graph by a user, wherein the average value of all the attributes of the brushed years is taken as the attribute of a single time point in the cross hysteresis path analysis model;

adjusting variables, wherein a user adjusts the transnational transfer variables participating in calculation in the cross hysteresis path analysis model by clicking a button at the upper left corner of the influence diagram;

expanding the comparison result, wherein a user clicks a button on the left side of the influence graph to expand the comparison result between different types of influences;

highlighting attributes the user clicks on the bar graph in the influence graph to highlight the bar graph at the same location each year for clear observation.

In the above-mentioned influence view, the year may be brushed in the influence view, and the average value of each attribute of the brushed year may be used as the attribute of a single time point in the cross-lag path analysis model; the cross country transfer variables participating in calculation in the cross hysteresis path analysis model can be adjusted by clicking a button affecting the upper left corner of the view; the button on the left side of the impact graph can be clicked to expand the comparison between the different types of impact (outlet or inlet); the bar graph in the impact graph may be clicked to highlight the bar graph at the same location each year for clear observation.

Preferably, the adjustment view includes an adjustment portion and a result portion, the adjustment portion including a slider corresponding to a selected transnational transition variable in the influence view, through which a user adjusts a player variable of the transnational transition; the results section contains text representing the exact value of the national team performance change.

The user can adjust the player variable of the transnational meeting at the adjusting part through the sliding block and view the adjusting result in real time. Two adjustment panels are placed in the adjustment view according to the target year and are arranged in time order to make a comparison between the two-way transnational adjustment of adjacent years. In this way, the user can compare the results of the adjustment panels to conclude that the year and direction of adjustment can significantly improve the performance of the national team.

There is a correlation between views in the visual analysis system above and a rich interaction mechanism is provided, including click, swipe, switch, box, mouse hover and drag slider operations.

For a click operation, the user may select an intercontinental football league in the map component, or a country in the similarity component, by clicking in the country view. In the effect view, the user can adjust the cross country transfer variables in the cross hysteresis path analysis model that are involved in the calculation by clicking on the button in the upper left corner of the effect view, and the bar graph in the effect view is clicked to highlight the bar graph at the same place every year for clear observation.

For the swipe operation, in the influence view, the user can swipe the year in the influence graph, and the average of the properties of the swiped year will be the property of a single point in time in the cross-lag path analysis model.

For the switching operation, in the country view, the user can switch whether to display an import player or an export player in the country view through a button. In the influence view, the user can click on a button on the left side of the influence graph to expand the comparison between different types of influences (outlet or inlet).

For the box selection operation, in the country view, the user can select a country in the similarity component through a mouse box selection, and can also remove the already selected country through a mouse box selection back selection.

For a mouse-over operation, in a country view, a user may obtain detailed information of map components and similarity components intercontinental football leagues or player-to-country transfer variables in a country by mouse-over. In the adjustment view, the user can view the exact value of the national cross-country transfer variable by hovering a mouse over a slider of the adjustment portion.

For drag slider operations, in a national view, a user may filter the time interval of interest by dragging a slider over the view. In adjusting the view, the user may adjust player variables of the transnational meeting in the adjustment section by dragging the slider.

In a second aspect, the method for visualizing the cross-country transfer analysis of the football player provided by the invention is realized based on a system for visualizing the cross-country transfer analysis of the football player, and comprises the following steps:

1) Acquiring the cross-country transfer data of players in the top-level football tournament of each country, namely the number of players, the price of the players, the average number of times of play of the single season of the players and the average number of goals of the single season of the players, and acquiring the international football rank points of each country;

2) Visualizing the nationwide transfer data of the football players obtained in the step 1) to obtain an overview of the nationwide transfer variables of the global football players, obtaining a dimension reduction projection result of the nationwide transfer variables of the football players in each country, selecting one country, and determining a data set for further analysis;

3) Calculating influence intensity by adopting a cross hysteresis path analysis model according to the country selected in the step 2), visualizing the influence intensity and displaying the influence intensity to an influence view, and carrying out detailed analysis on influence between the transition of the player across the country and the performance of the team of the country; simultaneously selecting the year, and adjusting the selected year in the adjustment view;

4) And according to the country selected in the step 2) and the year selected in the step 3), adjusting the transnational transfer variables in the adjustment view and obtaining the result of the adjusted national team performance.

Step 2) comprises:

2-1) dragging a sliding bar to select a cross-country transfer time period of the interested player;

2-2) clicking buttons corresponding to four transnational meeting variables to switch the variables of interest;

2-3) clicking buttons corresponding to player input and player output to switch the direction of the player transferring across the country;

2-4) adding the intercontinental football league or country of interest to the influence view by clicking or mouse-over, and removing the already selected country by mouse-over counter-selection.

The step 3) comprises the following steps:

3-1) clicking a cross hysteresis path analysis model selected by a player transnational transfer attribute to calculate an attribute adopted by the influence intensity;

3-2) observing the total influence intensity of the input player and the output player on the national team performance and comparing the result;

3-3) clicking an input player or output player button to check detailed influences and causal relations between the input player or output player and the national team performance;

3-4) clicking on the year tab selection year, and adjusting the selected year in the adjustment view.

Compared with the prior art, the invention has the following advantages:

according to the method for visualizing the cross-country transfer analysis of the football player, provided by the invention, the data analyst of the football player can easily, intuitively, comprehensively and deeply analyze the cross-country transfer policy of the football player by only importing the cross-country transfer data and the national team performance data of the football player in the corresponding format, so that the influence intensity between the cross-country transfer and the national team performance of the football player can be rapidly and clearly explored, and the cross-country transfer policy of the football player can be adjusted more specifically later. The field specialist gives a high degree of certainty to the invention.

Drawings

FIG. 1 is an interface schematic diagram of a system for cross-country transfer analysis visualization for soccer players in accordance with an embodiment of the present invention;

FIG. 2 is a schematic diagram of an influence diagram in an embodiment of the present invention.

Detailed Description

The present invention will be further described with reference to the following examples and drawings for the purpose of making the objects, technical solutions and advantages of the present invention more apparent. It will be apparent that the described embodiments are some, but not all, embodiments of the invention. All other embodiments, based on the described embodiments, which a person of ordinary skill in the art would obtain without inventive faculty, are within the scope of the invention.

Unless defined otherwise, technical or scientific terms used herein should be given the ordinary meaning as understood by one of ordinary skill in the art to which this invention belongs. As used in this specification, the word "comprising" or "comprises", and the like, means that the element or article preceding the word is meant to encompass the element or article listed thereafter and equivalents thereof without excluding other elements or articles. The terms "connected" or "connected," and the like, are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "upper", "lower", "left", "right", etc. are used merely to indicate relative positional relationships, which may also be changed when the absolute position of the object to be described is changed.

Examples

Referring to fig. 1, the system for cross country transfer analysis visualization of a soccer player of the present embodiment includes:

a country view, comprising a map component (as shown in A1 of fig. 1) and a similarity component (as shown in A2 of fig. 1), showing transnational transfer strategies for each tournament in the intercontinental football league and country/region, to facilitate browsing, selection and further analysis by the expert;

an influence view, providing a customized visualization component influence graph (shown as B in fig. 1), including total influence (shown as B2 in fig. 1) and detailed influence (shown as B3 in fig. 1), the expert can observe the change in the intensity of the bi-directional influence and further analyze the causal relationship;

adjusting the view (as shown in fig. 1C) provides an interactive method by which an expert can adjust the number and quality of players in a cross country and obtain simulation results.

Football analysts involved in the evaluation need to observe the impact of the player transnational transfer policies in the European football association (European football association) and provide a reference for further policy formulation. According to analysts, top-level tournaments in the European and soccer countries, particularly the five-major tournaments, are the most developed tournaments, attracting thousands of top-level players worldwide. The european union plays an important role in the global player's tendency to turn across countries. Therefore, analysts may be interested in the transition across countries of soccer countries/regions.

The map component in the country view displays a global player transnational meeting overview along with a statistical world map for analysts to view the intercontinental football league of interest (as shown in A1 of fig. 1). The analyst may switch the transnational transfer variables using buttons above the statistical world map. Opacity is used in the map component to encode particular values of variables of the import or export players for the selected country. The similarity component employs a scatter plot to display the dimension reduction projection results of all country transnational meeting variables (as shown in A2 of fig. 1). The opacity is used in the scatter plot to encode the transnational transition variables for each country. The selected countries in the map component and the similarity component are highlighted by adopting colors at the same time, and an analyst can observe the selected intercontinental football league or country at the same time in the two components. A slider moving on the view may filter the time period of interest. While the display of the entry player or the exit player at the overview assembly may be switched by a switch button. The analyst may click on the intercontinental football league of interest in the map component or box the country of interest in the similarity component for further detailed analysis.

First, the analyst adjusts the time slider to 1992 to 2018 and selects all countries/regions of the european union in the country view (as shown in A1 of fig. 1). The analyst then observes the values of the average import and export player variables over the years in the map component. From observation, the analyst found that the value of each import player variable was significantly higher in the european union country than in the other countries, and that the value of the export player variable was also higher in most other countries. Thus, the analyst is primarily concerned in further analysis about the impact between player importation and performance of the European Union team and adds all European Union countries/regions to the impact view.

The effect view contains a new design effect map to visualize the results of the cross-lag path analysis model (shown as B2, B3 in fig. 1). Comparing impact strengths by year is the primary task in the impact view. Thus, for direct comparison, the system places the bar graph directly with the timeline to show the impact strength over the years. The height of the bar graph represents the impact strength calculated by the cross-hysteresis path analysis model. In order to clearly show the trend of variation affecting intensity, an exponential scaling option of bar chart height is provided in addition to the linear scaling. The influence map can be switched between a total influence and a detailed influence. Under the total influence mode, an analyst can compare the total influence intensity of an import player and an export player and simultaneously find the change trend. In the detailed impact mode, the system uses circles to encode the primary impact direction between the player variable of the transnational meeting and the national team performance. The color of the circle codes the main direction of influence, while the area codes the difference in the intensity of influence in both directions.

Analysis of the export player from 2015 to 2018 is illustrated, for example, using an influence diagram in the detailed influence mode, as shown in fig. 2. The upper set of bar graphs represents the impact of export players on the performance of the national team, and the lower set represents the impact of the national team performance on the export players. Each column of the annual impact map contains four bar graphs. The four bar graphs represent the impact on player number, market value, number of plays and goals, respectively (as shown in fig. 2B, D). When analysts focus on the trend of variation in impact intensity between a particular transnational meeting variable and national team performance, they can compare bar graphs at the same location each year. To find the dominant direction of influence between the exporter and the national team performance, the analyst may observe a circle between the two bar graphs (as shown in fig. 2C). Through various comparisons, analysts may perform a comprehensive analysis of the player's transnational meeting.

After adding the European Union country/region to the influence view, the analyst begins to evaluate the effect of the important transnational transfer policies set forth between 1992 and 2018 (as shown in FIG. 1B). Analysts indicate that the cross-country transfer policy of players directly affects the number of cross-country transfer players and the market value. Thus, the analyst selects these two transnational transfer variables in the following analysis. In view of the delay in policy effects, analysts merge three or four consecutive years to more clearly observe their impact. The results are shown in the influence view in the form of a bar graph.

With respect to the total impact (as shown by B2 in fig. 1), the analyst found that the total impact from importer to national team performance gradually increased from 1996 to 2010, decreased slightly from 2011, and remained stable after 2011. This is consistent with the knowledge of the analyst. The analyst explained that this is due to the Bossman act in 1995 and the "native player rules" introduced in 2006. The bosman act, according to the analyst's theory, is to allow players to more freely transfer between clubs in europe and to promote the transfer of players across countries/regions in the european union. The European Union has proposed the "Ben Tuber rules" to force the number of Ben Tuber's players to play and further have adverse effects on players' transfer from other countries across to European Union countries. The analyst concludes that the bosman act significantly increases the impact of importation players on national team performance, while the "native player rules" reduces the intensity of this impact to some extent. Thus, an analyst analyzes the internal processes of the impact by observing detailed impact.

For detailed effects (shown as B3 in fig. 1), the analyst observes two cross-country transfer variables selected, respectively. To evaluate the impact of player numbers, analysts found that from 1996 to 1998, the number of import players was largely affected by national team performance, which is in line with the effects of the bosman act. The analyst explains that in the free-wheeling environment, foreign players are more likely to choose countries with higher levels of national teams to seek better development opportunities. The analyst then notices that from 1999 to 2007, importers have significantly more impact on national team performance than other impact directions. They explain that after a large-scale player transfer across the country, importers in the european football countries have a positive influence on the level of domestic players and further influence the performance of the national team. Meanwhile, with the transnational meeting of the players in the period, the number of imported players in other countries of the European and soccer is gradually increased, so that the influence of the national team performance on the number of imported players is reduced. The analyst also explained that the change in impact strength after 2008 corresponded to the impact of the "native player rules", which resulted in a decrease in the impact of national team performance on the number of imported players, while the impact in the other direction increased. To evaluate the impact of market value, analysts observed that from 2008 to 2010, import players had a very large impact on national team performance. This indicates that during this period, the player has an excessive price, which is effectively controlled by the European football-connected financial fair competition regulations proposed in 2010.

After selecting the target year in the influence view, the analyst may perform further simulation analysis in adjusting the view (as shown in fig. 1C). The analyst needs to perform a simulation analysis based on the path analysis model of the crossover lag. Analysts tend to adjust the number and quality of import and export players and then evaluate how the performance of the national team changes. Thus, an adjustment panel is provided in the influence view for an analyst to adjust the cross country transfer variables and directly analyze the results (as shown in fig. 1C 1, C2). The system focuses on adjusting the operation and the result display without exhibiting the simulation process. The adjustment panel may be divided into two parts, an adjustment part and a result part. The adjustment section includes a slider corresponding to a selected transnational transition variable in the influence view. The results section contains some text to represent the exact value of the national team performance change. The analyst can adjust the player variable of the national transfer in the adjustment part through the sliding block, and view the adjustment result in real time. Two adjustment panels are placed in the adjustment view according to the target year and are arranged in time order to make a comparison between the two-way transnational adjustment of adjacent years. In this way, the analyst can compare the resulting portions of the adjustment panel to conclude that the adjustment in which year and in which direction can significantly improve the performance of the national team.

In the tuning view, the analyst has selected a typical export-dominant national Argentina to verify his hypothesis (as shown in FIG. 1C). The analyst compares the import and export players of Argentina and found that the number and quality of export players exceeded those of the import player, which demonstrated Argentina to be the dominant country of export. The analyst then slightly increased the number of export players and noted that the international football score and ranking of Argentina also increased. Thus, the domain expert demonstrates that increasing the number of players to export can improve the national team performance in the country where player export is dominant.

In conclusion, the method for visualizing the cross-country transfer of football players can effectively and rapidly help users to explore the relationship between the cross-country transfer of football players and the national team performance, and support powerful multi-step analysis.

Claims

1. A system for cross country transfer analysis visualization for football players, comprising:

adjusting views, providing an interaction method for an expert to adjust the quantity and quality of the nationwide meeting players and obtaining a simulation result;

the interaction between views in the system comprises the following steps:

2. The system for visualizing a national transition analysis for a soccer player of claim 1, wherein said national view comprises a map component for displaying a global player national transition overview and a statistical world map and a projected similarity component for providing a global national transition strategy in terms of similarity.

3. The system for cross-country transfer analysis visualization of football players of claim 2, wherein the country view comprises the following interactions:

4. The system for cross-country transfer analysis visualization of soccer player of claim 1, wherein the impact view comprises an impact graph that places a bar graph with a timeline to show impact strength over years; the height of the bar graph represents the impact strength calculated by the cross-hysteresis path analysis model.

5. The system for cross-country turn analysis visualization of football players of claim 4, wherein in the detailed influence mode, circles are used to encode the dominant directions of influence between player variables and the national team performance of the cross-country turn, the colors of the circles encode the dominant directions of influence, and the areas encode the differences in the intensities of influence of the two directions.

6. The system for cross-country transfer analysis visualization of soccer player of claim 4, wherein the influence view comprises the following interactions:

7. The system for visualizing a national transition analysis for a soccer player according to claim 1, wherein said adjustment view comprises an adjustment portion and a result portion, said adjustment portion comprising a slider corresponding to a selected national transition variable in said influence view, whereby a user adjusts a player variable of a national transition; the results section contains text representing the exact value of the national team performance change.

8. The system for cross country transfer analysis visualization of a soccer player of claim 1, wherein step 2) comprises:

9. The system for cross country transfer analysis visualization of a soccer player of claim 1, wherein step 3) comprises: